exons & protein domains

[Geoff Barton]

Andrew J. Roger wrote:
> 
> Gary DosSantos wrote:
> >
> > This is off the topic, but I was wondering if any algorithms have been
> > written that allow for the generation of phylogenetic trees by aligning
> > structural domains. A typical protein does not have all that many
> > structural domains, certainly not enough to perform a bootstrap
> > analysis. It seems to me then, that by using structural domains as a
> > criteria for protein alignment, one could not obtain the amount of data
> > necessary for estimating the degree of relatedness between groups of
> > proteins, but could only say that such and such proteins are related.
> 
> Look in the last chapter of Methods in Enzymology, Volume 183. There
> is a paper using three dimensional structure to construct
> evolutionary trees (it is a distance method, I believe).
> 

Yes.  That is a good start, but there has been a lot of work done 
over the last 5 years on the subject of comparing protein 3D structures
using a variety of different measures of similarity or distance.  
Look at the work of Taylor and Orengo and Sali and Blundell and Holm 
and Sander for a start.

Although similarities of protein three dimensional structure are often 
used to infer that the two proteins have a common ancestor.   On
its own, structural similarity is rarely enough to confirm homology.

For an intriguing example from my own work see Russell and Barton
(1993), Nature, 364, 765.

Geoff. 

-- 
Geoffrey J. Barton, Laboratory of Molecular Biophysics, University of
Oxford, 
Rex Richards Building, South Parks Road, Oxford OX1 3QU, U.K.
mailto:gjb at bioch.ox.ac.uk, Tel: +44 1865 275368, Fax: +44 1865 510454, 
ftp://geoff.biop.ox.ac.uk, http://geoff.biop.ox.ac.uk